Finalizing an XGBoost model

It's time to build a robust XGBoost model to add to the pipeline. Go ahead and import XGBRegressor, numpy, GridSearchCV, cross_val_score, KFold, and mean_squared_error as follows:

import numpy as np
from sklearn.model_selection import GridSearchCV
from sklearn.model_selection import cross_val_score, KFold
from sklearn.metrics import mean_squared_error as MSE
from xgboost import XGBRegressor

Now let's build the model.

First XGBoost model

This Student Performance dataset has an interesting range of values for the predictor column, y_train, which can be shown as follows:

y_train.value_counts()

The result is this:

11    82
10    75
13    58
12    53
14    42
15    36
9     29
16    27
8     26
17    24
18   ...